The present invention pertains to methods and compositions providing rapid, sensitive and specific detection of chemical warfare nerve agents. Chemical warfare agents are divided into a number of classes, the most prominent of which are vesicants and nerve agents. The former category includes sulfur and nitrogen “mustards” that typically incorporate 2-haloethylsulfide or 2-haloethylamine units. These compounds are often termed “blistering” agents, and are both strong alkylating agents and sources of localized hydrochloric acid. Incapacitation (and eventual death) may occur with topical exposure, but they are most effective when inhaled, due to rapid lung damage.
Though vesicants were used with horrifying effect in the First World War and more recently in Iraq, their utility as weapons and terror agents pale in significance when compared to nerve agents. Scheme 1 provides structures of several specific nerve agents and a generalized structure typical of all common nerve agents. These compounds, exemplified by sarin (isopropyl methylfluorophosphonate), share a common structural motif: they are esters of a methyl phosphonic acid derivative that incorporates a moderately good leaving group at phosphorus. It is important that the leaving group is only moderately good, as premature hydrolysis by water will occur with good leaving groups (e.g., chloride, bromide). The utility of nerve agents stems from their extraordinary toxicity; a lethal dose can be as little as 0.01 mg/kg in man (<2 mg for most adults).

Nerve agents act by inhibiting the enzyme acetylcholinesterase, and their design can be understood on this basis. Muscular contraction is initiated by release of the neurotransmitter acetylcholine across a synaptic junction. If the acetylcholine is not somehow removed, a continuous Aon@ signal to the muscles will be delivered, resulting in paralysis. Acetylcholinesterase accomplishes this removal by hydrolyzing the ester linkage to give acetate and choline (which is later recycled). The acetylcholine is bound in an active site that includes a hydrophobic pocket for the ethyltrimethylammonium group and a hydrogen bond donating group for the acetyl carbonyl. Attack by an active site serine residue leads to a key tetrahedral intermediate that breaks down to give an acetylated serine residue and choline. Loss of choline is followed by enzymatically mediated attack by water as a nucleophile, to again form a tetrahedral intermediate that breaks down to give acetate and the original serine residue. The key catalytic action of the enzyme involves facilitating the formation of tetrahedral intermediates involving the active site serine residue.
Nerve agents combine the features of active site directed reactive agents and transition state analogs. Binding of the nerve agent in the active site of acetylcholinesterase is enhanced by interaction of the alkyl group of the ester with the hydrophobic binding pocket of the enzyme, and by the interaction of the phosphoryl oxygen with a hydrogen bonding group. Reaction with the serine hydroxyl gives a phosphonate ester. Importantly, this phosphonate ester closely resembles the tetrahedral intermediate that the enzyme is designed to stabilize. Though the carbon based tetrahedral intermediate formed in the natural enzymic reaction is inherently unstable, this is not the case for the phosphonate ester; this serine phosphonate ester is quite stable, and is not subject to ready hydrolysis. As a result, the enzyme is left in a condition in which the serine residue is tied up as an ester, with the active site blocked. No further catalytic action can take place, and any acetylcholine released by nerve cells is not degraded, leading to a permanent “on” state of muscular contraction. From this description, it may be seen that the essential features of a nerve agent are simply a tetrahedral phosphorous having a P═O bond and a leaving group, leading to the generalized formula: XYP(O)MLG (see, Scheme 1 for the corresponding generalized structure). The other groups V and L attached to the phosphorous should not be so large as to interfere with binding to the enzyme, but there is considerable latitude with respect to the actual identities; not only may their size vary, but they may be alkyl, ester or dialkylamide groups (e.g., Tabun).
The threat posed by nerve agents has made them important targets for detection. Analytical methods employed have included gas-chromatography-mass spectrometry, HPLC methods with various detection systems (e.g. UV, refractive index, post-column derivatization) and capillary electrophoresis. More novel methods include those based on biosensors and immunoassays. Many of these methods are excellent for the purpose of retrospective analysis; that is, for examining a site and determining whether nerve agents have been used recently or relatively long ago (i.e., methods that focus on detecting nerve agent degradation products). However, most of these methods require instrumentation that is emphatically non-portable, quite expensive, and which is not well suited to the rapid analysis of multiple samples—or the even more attractive goal of continuous or near-continuous real-time analysis.
All of the analytical methods described above share an additional disadvantage: they are specific to the molecular identity of the nerve agents subject to detection. Analytical techniques typically begin with the separation of the components of a sample matrix (e.g., by HPLC, GC, TLC, CE). In the simplest analytical methods identification is made on the basis of elution time (or elution distance, in the case of TLC), while more sophisticated methods couple this time information with spectral identification (e.g., molecular mass, redox potential, etc.). Unfortunately, the very specificity of these methods may lead to a false sense of security; if either the elution time of a substance or the characteristic spectral signature of the compound is different than what has been defined for a target nerve agent, then the compound will not be identified as a threat. Yet, it may still be a threat. A somewhat facetious example of this would be the use of d3-sarin, CD3P(O)(O-iPr)F. When analyzed by GC-MS, this compound would have an essentially identical retention time as sarin, but would not be identified as such, since its mass would differ; it would simply be considered to be some harmless, co-eluting impurity. A more realistic example would be the use of a less common ester of a methylfluorophosphonate (e.g., CH3P(O)(OEt)F), or a phosphinate ester (e.g., CH3iBuP(O)F, the isostere of sarin); while it is true that compounds like these will not be quite as effective as sarin or soman, they will nevertheless provide a high killing rate.
As pointed out above, the essential features of a nerve agent are a tetrahedral phosphorus with a P═O bond and a moderately good leaving group; one need only look at the similar effectiveness of Sarin, Soman and Tabun and their structural diversity with respect to leaving group and other phosphorous substituents (See, e.g. Scheme 1) to realize that simple permutations around phosphorous will lead to a multitude of nerve agents, all having unique retention times by various separations methods, and all having unique spectral signatures. Some of the analytical methods described above could be adapted to the detection of this multitude of compounds (e.g., GC-MS) at the cost of developing and maintaining an extensive and sophisticated library of retention times and spectral signatures. However, in many of the methods (e.g., the highly specific immunoassays), even this would be impractical; scores of individual tests would have to be run.
In sum, the bulk of the analytical methods available for nerve agents are politically useful, for the purpose of determining whether an enemy (or potential enemy) has tested or used specific, known nerve agents. However, the bulk and power requirements of most of the instruments employed make them impractical for battlefield conditions, and in the rare cases where this is not a problem, they are often incapable of detecting the presence of substances that can act as nerve agents rapidly enough to avoid exposure. Finally, these analytical methods share the common limitation that they can only detect compounds that are expected, whereas in a practical sense it doesn't matter if troops are going to be killed by the most sophisticated and “popular” nerve agent available; what matters is that they are being exposed to a rapidly acting acetylcholinesterase inhibitor, and that substance may differ from those commonly detected through convenience (i.e., ready manufacture), or more chillingly, by purposeful design to elude standard detection methods. Clearly, it would be desirable to have an analytical method that could rapidly detect substances that can act as nerve agents regardless of their exact structures, and which could accomplish this feat using instrumentation that was compact, robust and easily powered—or utilize no instrumentation at all.